Creator

Google DeepMind

Released

2025-06-05

Intelligence

34.6

Artificial Analysis Index

Coding

32.0

Artificial Analysis Index

In $/1M

$1.25

input tokens

Out $/1M

$10.00

output tokens

Blended $/1M

$3.44

3:1 blended

Speed

139

tokens / sec

Benchmark breakdown

Independent evaluation scores, normalized to 0–100.

GPQA Diamond
84
Humanity’s Last Exam
21
MMLU-Pro
86
SciCode
43
LiveCodeBench
80
MATH-500
97
AIME 2025
88
τ²-Bench (agentic)
54
Terminal-Bench Hard
27
IFBench
49
Via Artificial Analysis